1. Field of the Invention
The present invention relates to an image forming apparatus which forms an image with pixels, such as a digital copying machine, a digital printer, an ink jet printer and a printing machine, and in particular relates to an image forming apparatus which embeds information, which is different from an image visually recognized and outputted as a visual image, in the image area or its background area so that the embedded information may not be recognized.
2. Discussion of the Related Art
In the field of image forming, addition of information to an image, which is different from the image such as a text or characters, has been conventionally conducted for increasing the information amount, keeping the information secret from a third party, and so forth.
For example, Japanese Patent Application Laid-Open No. Hei. 7-123244 (1995) discloses an invention providing an image processing apparatus which embeds information such as text information which is different from the color image to be formed in the process of forming a color image by means of changing the value of either color difference or chroma so that the total amount of three primary color components of the color image does not vary. That is, the invention embeds the different information in the color image so that the information may not be recognized utilizing the property of color difference and chroma that the variations in their high frequencies are hardly recognized by the user.
However, in this invention, the information can be embedded only in the position where the color image exists; therefore, the available areas for information embedding in the image are limited, and besides, change in the value of color difference or chroma should be considerably large so as to read the embedded information securely by a scanner or the like, and thereby the quality of the original color image is deteriorated.
Japanese Patent Application Laid-Open No. Hei 4-294682 (1992) discloses an image processing apparatus embedding the product number, the user identifier and so forth in the color image using yellow toner with a specific pattern which is difficult to be identified by user's observation. That is, this invention embeds the information different from a color image in the color image so that the information may not be recognized utilizing the user's observation characterized in that it is difficult to identify a specific pattern formed by the yellow toner. However, the invention has a default that the specific pattern is recognized in the portion of low density in the image, and thereby the quality of the color image is deteriorated.
Japanese Patent Application Laid-Open No. Hei. 5-301380 (1993) discloses an image processing method which examines the density of a color image, and according to the result, adds no specific pattern to the portion of low image density. This invention is able to overcome the above-described default, but there is another problem such that the area available for embedding the specific pattern is limited, and accordingly, the amount of information to be added is decreased.
Japanese Patent Application No. Hei. 4-334266 (1992) discloses methods and means for embedding machine readable digital data in halftone images utilizing digital data in the shapes of the marks called "glyphs". However, this invention has some problems such that the halftone image must be formed in the original image, and that the amount of the additional information is small because the marks constituting glyphs themselves are larger than pixels.
Japanese Patent Application Laid-Open No. Hei. 6-113111 (1994) discloses an image processing apparatus by which two states of color image data are formed. These two states of data have different amounts of K (black) by which a part of Y (yellow), M (magenta) and C (cyan) is replaced so that these two states of data may result in the same color image, and ID information is added to the image by assigning 0 or 1 to each of these states. However, the ID information cannot be added to a monochrome image, and besides, in some situations, two states cannot be formed for all of pixels in the image; therefore, there is a possibility of causing a case where it is impossible to add ID information.
Japanese Patent Application Laid-Open No. Sho. 63-214067 (1988) discloses a data embedding method for embedding information different from an image in the image by determining arrangement of elements in a dither matrix according to the information to be added in a process of representing the image by the dither method. Japanese Patent Application No. Hei. 2-266390 (1990) discloses an improvement of the above-described data embedding method which embeds information different from an image in the image by assigning 0 and 1 to the outputs of pairs of threshold value difference k (1, 0) and (0, 1) by application of the dither matrix.
However, these inventions using dither matrix have defaults such that the different information can be embedded only in a specific portion of the image, and that it can be applied only to the image forming apparatus employing the dither matrix as the representation of gradations. Furthermore, in the case where the information is embedded in an electronic image, it is easy to read the embedded information later because the position of the dither matrix can be specified, but in the case of the image outputted by the printer, it is almost impossible to specify the two dimensional position or pattern of the dither matrix in the image scanned by an image scanner.
As described above, various kinds of inventions for embedding and adding information different from an image to the image have been proposed, but there have been some problems such as the small amount of information which can be added, deterioration of image quality caused by addition of information, or difficulty in reading the embedded information from the image.